Reviewers ask for analyses that don't fit. Study sections doubt the power analysis. Models fail to converge, and results contradict the literature. Each Case in the Clinic explains what the symptom means, what usually causes it, which checks to run, and how to explain the decision.
Pick the description closest to where you're stuck. Each department leads to focused Cases rather than a general methods archive.
Requests for post hoc power, multiple-comparisons corrections, different models, or more covariates, and how to answer them without weakening the paper.
Effect sizes from pilots, unknown ICCs, attrition, and the sample size justification a study section will read closely.
Convergence failures, singular fits, and estimation problems in the multilevel, longitudinal, and latent variable models research teams rely on.
Effects that flip sign, vanish, or contradict the literature, and how to tell a real finding from a modeling artifact.
Missing waves, uneven clusters, attrition, and measurement problems that change what the analysis can support.
A coauthor reruns the code and gets different numbers. Finding where the analyses diverged, and preventing it next time.
Why "observed power" can't answer the reviewer's real question, and the three analyses that can.
Whether to correct depends on the claim you're making, not on how many p-values are in the paper.
Small pilots give effect sizes too noisy to plan on, and the pilots that look most promising are the most misleading.
A quick fix can make a warning disappear or a comment go away while leaving the inference weaker than before. Each Case explains what the problem means for your conclusions, compares the defensible responses, and gives you wording you can adapt. Every number comes from an R script you can download and run.
Working on a dissertation? Dissertation Stats Helper's Analysis Clinic covers the problems doctoral students meet most. Preparing a manuscript? Reviewer Ready reviews the statistics before you submit.
Upcoming Cases become available once the diagnosis, the reproducible example, and the response guidance have been checked.
Decide whether clustering changes your inference, and answer with evidence rather than a cutoff.
Build the power analysis around a defensible range of ICCs instead of a single guess.
Separate optimizer trouble from a random-effects structure the data cannot support.
Tell suppression, confounding, and collinearity apart before you interpret the new sign.
Match the missing-data method to the design and state the assumption it rests on.
Trace the difference to data versions, defaults, seeds, or package changes.
Some problems turn on the details of your design, data, or the exact reviewer comment. A free 30-minute consult can identify the next defensible step and what it would take.
Book a free consult